linux-kernel - [PATCH 5.10 139/306] bus: fsl-mc: fix mmio base address for child DPRCs

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20210916155758.780448513@linuxfoundation.org>
Date:   Thu, 16 Sep 2021 17:58:04 +0200
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     linux-kernel@...r.kernel.org
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        stable@...r.kernel.org, Laurentiu Tudor <laurentiu.tudor@....com>,
        Sasha Levin <sashal@...nel.org>
Subject: [PATCH 5.10 139/306] bus: fsl-mc: fix mmio base address for child DPRCs

From: Laurentiu Tudor <laurentiu.tudor@....com>

[ Upstream commit 8990f96a012f42543005b07d9e482694192e9309 ]

Some versions of the MC firmware wrongly report 0 for register base
address of the DPMCP associated with child DPRC objects thus rendering
them unusable. This is particularly troublesome in ACPI boot scenarios
where the legacy way of extracting this base address from the device
tree does not apply.
Given that DPMCPs share the same base address, workaround this by using
the base address extracted from the root DPRC container.

Signed-off-by: Laurentiu Tudor <laurentiu.tudor@....com>
Link: https://lore.kernel.org/r/20210715140718.8513-8-laurentiu.tudor@nxp.com
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Signed-off-by: Sasha Levin <sashal@...nel.org>
---
 drivers/bus/fsl-mc/fsl-mc-bus.c | 24 ++++++++++++++++++++++--
 1 file changed, 22 insertions(+), 2 deletions(-)

diff --git a/drivers/bus/fsl-mc/fsl-mc-bus.c b/drivers/bus/fsl-mc/fsl-mc-bus.c
index 806766b1b45f..e329cdd7156c 100644
--- a/drivers/bus/fsl-mc/fsl-mc-bus.c
+++ b/drivers/bus/fsl-mc/fsl-mc-bus.c
@@ -64,6 +64,8 @@ struct fsl_mc_addr_translation_range {
 #define MC_FAPR_PL	BIT(18)
 #define MC_FAPR_BMT	BIT(17)
 
+static phys_addr_t mc_portal_base_phys_addr;
+
 /**
  * fsl_mc_bus_match - device to driver matching callback
  * @dev: the fsl-mc device to match against
@@ -597,14 +599,30 @@ static int fsl_mc_device_get_mmio_regions(struct fsl_mc_device *mc_dev,
 		 * If base address is in the region_desc use it otherwise
 		 * revert to old mechanism
 		 */
-		if (region_desc.base_address)
+		if (region_desc.base_address) {
 			regions[i].start = region_desc.base_address +
 						region_desc.base_offset;
-		else
+		} else {
 			error = translate_mc_addr(mc_dev, mc_region_type,
 					  region_desc.base_offset,
 					  &regions[i].start);
 
+			/*
+			 * Some versions of the MC firmware wrongly report
+			 * 0 for register base address of the DPMCP associated
+			 * with child DPRC objects thus rendering them unusable.
+			 * This is particularly troublesome in ACPI boot
+			 * scenarios where the legacy way of extracting this
+			 * base address from the device tree does not apply.
+			 * Given that DPMCPs share the same base address,
+			 * workaround this by using the base address extracted
+			 * from the root DPRC container.
+			 */
+			if (is_fsl_mc_bus_dprc(mc_dev) &&
+			    regions[i].start == region_desc.base_offset)
+				regions[i].start += mc_portal_base_phys_addr;
+		}
+
 		if (error < 0) {
 			dev_err(parent_dev,
 				"Invalid MC offset: %#x (for %s.%d\'s region %d)\n",
@@ -996,6 +1014,8 @@ static int fsl_mc_bus_probe(struct platform_device *pdev)
 	plat_res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	mc_portal_phys_addr = plat_res->start;
 	mc_portal_size = resource_size(plat_res);
+	mc_portal_base_phys_addr = mc_portal_phys_addr & ~0x3ffffff;
+
 	error = fsl_create_mc_io(&pdev->dev, mc_portal_phys_addr,
 				 mc_portal_size, NULL,
 				 FSL_MC_IO_ATOMIC_CONTEXT_PORTAL, &mc_io);
-- 
2.30.2